The choice of iron fortificant usually represents a balance between bioavailability of the compound and its tendency to cause organoleptic problems. The aim of this study was to evaluate the efficacy of sodium iron EDTA (NaFeEDTA) and ferrous fumarate at levels compatible with South African brown bread (10 mg/kg flour for NaFeEDTA and 20 mg/kg flour for ferrous fumarate) in a randomized controlled trial; electrolytic iron was evaluated at the level currently used in South Africa (35 mg/kg flour). Schoolchildren (n = 361), aged 6-11 y, from a low socioeconomic community with hemoglobin (Hb) < or = 125 g/L were randomly assigned to 1 of 4 groups that received 4 slices of brown bread supplying either: 1) no fortification iron 2) 2.35 mg iron as NaFeEDTA; 3) 4.70 mg iron as ferrous fumarate; and 4) 8.30 mg iron as electrolytic iron per intervention day. These amounts simulated a bread intake of 6 slices per day over the 34-wk study period at fortification levels of 0, 10, 20, and 35 mg/kg flour, respectively. Hb concentration and iron status were assessed at baseline and after 34 wk of intervention. The iron interventions did not affect Hb concentration, transferrin saturation, or serum ferritin, iron, or transferrin receptor concentrations relative to the control group. Our results suggest that electrolytic iron at the level currently used in South Africa is not effective in improving iron or Hb status. Neither do NaFeEDTA or ferrous fumarate appear to be suitable alternatives for the fortification of wheat flour when included at levels that do not cause color changes.